A conventional application for text processing typically requires specific knowledge of the implementation of text objects and their containers for performing editing functions to manipulate text objects and for performing layout functions for displaying the text objects. Usually the application includes code such as a text editor or a layout editor for arranging text objects for visual display. To function properly, the layout editor may require specific knowledge of the container to properly present the text objects within the container. Likewise, a text editor may require specific knowledge of the text objects to properly edit them. These editors typically perform a mapping function between content and layout.
It is easier to map between content and layout when each element of content is represented by one piece of layout. This may normally happen with forms where, for instance, one control may be represented by one button. However, this may not be the case for flowing content where one element of content may appear in many places in layout. For example, the content may flow around some complex shape or image. The relationship between structure of data and structure of layout may quickly become very complicated. Conventional editing operations thus rely on specific knowledge of the implementation of the text objects and container to efficiently translate between user view and the data representation of the content. Often these editing operations manipulate a whole element of content and move it within a given data structure such as deleting the element from or inserting the element in the data structure.
But where the details of the implementation data structure become hidden to the text processing application such that the text processing application does not have specific knowledge of the implementation of the text objects or their containers, then new editing techniques may be required for manipulating a representation of text objects and their containers. What is needed is a way for a text processing application to edit a data abstraction containing flowing content. Such a system and method should not require specific knowledge of the implementation data structure of text objects or their containers in order to perform editing operations. Moreover, editing operations for flowing content should be allowed to manipulate subelements of the content stream.